I have a pair of Heathkit AA-161 mono integrated amps using EL84/6BQ5/7189 in ultra-linear push-pull, with 6AN8 as driver, 6AU6 as line stage, and 6EU7 as phono stage, as illustrated in the schematic below. I have worked on other Heathkit models using 6BQ5 such as UA-2, AA-151, EA-2, EA-3, before and sometimes have reconfigured them radically using different driver circuits. But this time I want to keep it simple and closer to stock form. It's a work in progress. Nothing urgent and just want to have some fun with it. Currently one amp has more hum than the other, and it will need filter cap replacement, etc...
As many of you already know that with vintage gears is that modern household AC is typically 120vac or higher so when a vintage piece power transformer's primary was designed for typically between 105vac to 110vac so the extra voltage creates problems for amps that was already designed to push the tubes on the edge, maxing the total plate dissipation. I like to run my tubes conservatively and they sound more relaxed that way, too. I am not a big fan of ultra-linear at high voltage when the screen-grid is the same or higher than the plate voltage, leading to shortened tube life and tense sound when tube is pushed to the edge.
Long story short, I want suggestions on elegant ways to reduce the high voltage, either on the B+ or at the power transformer. Obviously I don't want to have to use a variac all the time when powering up these puppies.
Here is a stock schematic:
Here is the modded schematic so far. Not using half of 6EU7 as part of the line stage, only using the 6AU6. Phono using both halves of 6EU7. Using "Tuner" and "Crystal Phono" as line inputs, deleting all the junk in the signal path before going to the volume control. Other than that, everything is still in stock form. I like the tone controls so I would like to keep them if quiet enough as they can add some flavor to the sound. (Yes, I like coloration and tone so no lecture on "neutrality," please.)
When slam with 120vac the supposedly 390vdc B+ goes all the way to 430vdc! Not good.
.
As many of you already know that with vintage gears is that modern household AC is typically 120vac or higher so when a vintage piece power transformer's primary was designed for typically between 105vac to 110vac so the extra voltage creates problems for amps that was already designed to push the tubes on the edge, maxing the total plate dissipation. I like to run my tubes conservatively and they sound more relaxed that way, too. I am not a big fan of ultra-linear at high voltage when the screen-grid is the same or higher than the plate voltage, leading to shortened tube life and tense sound when tube is pushed to the edge.
Long story short, I want suggestions on elegant ways to reduce the high voltage, either on the B+ or at the power transformer. Obviously I don't want to have to use a variac all the time when powering up these puppies.
Here is a stock schematic:
An externally hosted image should be here but it was not working when we last tested it.
Here is the modded schematic so far. Not using half of 6EU7 as part of the line stage, only using the 6AU6. Phono using both halves of 6EU7. Using "Tuner" and "Crystal Phono" as line inputs, deleting all the junk in the signal path before going to the volume control. Other than that, everything is still in stock form. I like the tone controls so I would like to keep them if quiet enough as they can add some flavor to the sound. (Yes, I like coloration and tone so no lecture on "neutrality," please.)
An externally hosted image should be here but it was not working when we last tested it.
When slam with 120vac the supposedly 390vdc B+ goes all the way to 430vdc! Not good.
An externally hosted image should be here but it was not working when we last tested it.
.
One classic solution is called a "bucking" transformer, typically a filament transformer with its secondary in series with the Heathkit's primary. With the right polarity (check carefully!) the Heathkit will see 6 or 12 or whatever volts less.
This could be done safely in a separate box with a cord for the male end and a socket for the female end. 125 volts in, 115 volts out, more or less.
All good fortune,
Chris
This could be done safely in a separate box with a cord for the male end and a socket for the female end. 125 volts in, 115 volts out, more or less.
All good fortune,
Chris
Long story short, the Variac is precisely the best solution and it's not at all obvious that you don't want to do that. Quite the opposite.
You set the Variac to the voltage of your choice and leave it there. You DON'T play with it every time. Simple, infinitely adjustable, doesn't waste power as heat (like a resistor would) cheap and you don't have to modify the amp.
The problem is that if I loan them to friends who don't have a variac or use them at a different location, I don't want to haul around a variac all the time. I only have one variac and I need that for other projects. Long story short, using a variac at ALL time is not practical, at least for me. Thanks for the advice.
Also thanks for other members' advices above.
I'd do the bucking transformer too. It will lower the B+ and filament. On some of these older amps, now a days you'll find the filaments are running in excess of 7 volts. With the bucking transformer, the power transformer will run cooler, the bucking transformer will just get warm, and they are usually cheap, 12.6 volts at 1.2 amps at radio shack is $9.49. RS's 3 amp is $10.99. 12.6 volts will drop voltages inside 10 - 11%; 400 volt B+ is now 360, 7 volt filament is now 6.3. For less drop use a 6.3 volt transformer instead of a 12.6.
You don't want to put a resistor in there of 21 ohms or more as you will raise the DCR of that nice PS hurting the sonics. A choke of 21 ohms would be better, and Edcor has affordable ones in there new series.
Tube DIY Asylum
Another solution is to convert to fixed bias using Dave Gillespie's EFB mod. You can set the bias to 25-27 mA for long tube life. It only cost $10 to implement or you could do it with the conventional small add on transformer for fairly low $ .
The 5th post down has a link to his paper.
diytube.com :: View topic - Improved SCA-35/ST-35 Performance
Biasing
I agree the bucking trans would be better than a big resistor too.
Add ICL's too the input of PS trans too.
Those output transformers are the same on the UA amps and the AA-151 right?
Randy
Tube DIY Asylum
Another solution is to convert to fixed bias using Dave Gillespie's EFB mod. You can set the bias to 25-27 mA for long tube life. It only cost $10 to implement or you could do it with the conventional small add on transformer for fairly low $ .
The 5th post down has a link to his paper.
diytube.com :: View topic - Improved SCA-35/ST-35 Performance
Biasing
I agree the bucking trans would be better than a big resistor too.
Add ICL's too the input of PS trans too.
Those output transformers are the same on the UA amps and the AA-151 right?
Randy
Last edited:
I think a bucking transformer is probably the best route. This way, I can use the extra winding for, perhaps, fixed bias supply if I ever decided to go that route.
Thanks!
What is "ICL"? Sorry, I am not good at abbreviations.
Yes, same output transformer (51-29), also the following amps:
mono amps, power transformer (54-63)
UA-1
UA-2
EA-2
EA-3
AA-161
stereo amps, power transformer (54-93)
AA-111
AA-151
SA-2
All of these amps use similar circuit, ultra-linear, 6AN8 driver, etc...
.
ICL is a input current limiter. Mostly we use GE's CL series. You can find them at mouser, look for CL-90. An input current limiter has a high resistance when cold and much lower resistance when hot, it gets hot when running. Like a CL-80 is 50 ohms when cold, or unit is off. When first turned on, you have 50 ohms series resistance on the incoming AC, in 30 seconds or so the resistance has come down to something like 3 ohms. So when hot a CL-80 will drop something like 3 volts AC. The filaments, and rectifying diodes won't get a high surge current through them.
directdriver - I'm wondering how you made out with your mods? I have an AA-13 I'm going to re-cap. I've included a schematic. I did a quick comparison of the AA-161 vs AA-13 schematics and other than C34 and C35, and larger PS can cap values they seem identical. What I find strange is that both have large differences in the voltages shown on the schematics. Both use the same power and output transformers.
I plan on using a bucking transformer to reduce the voltage.
I plan on using the MM phono and tuner inputs for their intended purpose, but would like to use the xtal input for a cd player.
Will I realize better bass response if I increase the value of C18 and C19? Can you suggest which resistors to change to metal film. I will check the remaining carbon resistors for accuracy.
Thank you.
I plan on using a bucking transformer to reduce the voltage.
I plan on using the MM phono and tuner inputs for their intended purpose, but would like to use the xtal input for a cd player.
Will I realize better bass response if I increase the value of C18 and C19? Can you suggest which resistors to change to metal film. I will check the remaining carbon resistors for accuracy.
Thank you.
Increasing the value of C18/C19 is risky because they set the low frequency dominant pole rolloff in the feedback loop. The capability to amplify infrasonics isn't really a virtue, and recovery time from overload suffers linearly with this time constant. It's important to remember that the output transformer's low frequency pole shifts around with signal level and so dynamic stability can suffer with poorly conceived changes to the existing loop.
All good fortune,
Chris
Thx for the heads up on that Chris.
I'm new to Heathkit, are there any other mods or upgrades that would be beneficial to improving the sound or safety?
Can the tone controls be bypassed?
Can the pre-amp section be bypassed? I was thinking that would give me greater input options.
I'm new to Heathkit, are there any other mods or upgrades that would be beneficial to improving the sound or safety?
Can the tone controls be bypassed?
Can the pre-amp section be bypassed? I was thinking that would give me greater input options.
Last edited:
Sorry for the lack of update on this project. I've been busy with other projects.
Thanks for the schematic, TMS. I tried the bucking transformer and it turns out the input voltage is actually accurate. If I use the bucking transformer, the filament voltage becomes too low. The problem is the voltage after rectification and what's interesting is that it matches the voltage on the AA-13! The AA-13's operation points on the schematic are more accurately reflects what's on the AA-161's actual voltage, which is 415 to 425vDC on the B+. The only difference between the two is that the AA-13 allows 240vac connection, with two primary windings. It appears the AA-13 is probably the last version of 6BQ5 mono amps dating back to the EA-2.
I ended up grudgingly using a 1K 10-watt resistor in series with the B+ supply to knock down about 50 volts to be at around 365v. (If there's a simpler and more elegant way, please let me know) I also added an extra 22uf/450v decoupling cap. So after the 6CA4 rectifier tube is cap-1KΩ-cap and then B+. I changed the 210Ω cathode resistor to 270Ω to lessen the current. The tubes running less hot and the sound is more relaxed. I also use 56v zener diodes on the screen taps to drop the screen voltage to around 300v.
I actually like the sound and flexibility of the tone control, a little boost here and a little cut there, makes life lot easier. The bass itself is quite deep and awesome without increasing cap size. Unless you have leakage from the coupling caps, there's no need to change to fancy modern caps. They sound just fine. (I like coloration and I am sure the "neutrality" fascists will be all over me but I couldn't care less. Your mileage may vary.)
I have several of these amps so when I have the time I would like to use a pair to do a mod with completely different circuit. One day I would like to implement the Pilot 232 circuit on this amp.
Oh, I also change the 6AU6 to triode connection with a 3.3k cathode resistor and it sounds great, albeit with less gain.
I tried the phono input using two halves of the 6EU7 but the hum is too loud to be usable due to the AC filament. The phono sounds is quite nice but the hum is too loud. You need DC filament to be quiet enough. I just don't use it right now for this mod.
I also added 1kΩ grid resistors for all the tubes to avoid oscillation.
--------------------------------------------------------------------
Below are stock schematics of the two amps for comparison.
AA-161
AA-13
Thanks for the schematic, TMS. I tried the bucking transformer and it turns out the input voltage is actually accurate. If I use the bucking transformer, the filament voltage becomes too low. The problem is the voltage after rectification and what's interesting is that it matches the voltage on the AA-13! The AA-13's operation points on the schematic are more accurately reflects what's on the AA-161's actual voltage, which is 415 to 425vDC on the B+. The only difference between the two is that the AA-13 allows 240vac connection, with two primary windings. It appears the AA-13 is probably the last version of 6BQ5 mono amps dating back to the EA-2.
I ended up grudgingly using a 1K 10-watt resistor in series with the B+ supply to knock down about 50 volts to be at around 365v. (If there's a simpler and more elegant way, please let me know) I also added an extra 22uf/450v decoupling cap. So after the 6CA4 rectifier tube is cap-1KΩ-cap and then B+. I changed the 210Ω cathode resistor to 270Ω to lessen the current. The tubes running less hot and the sound is more relaxed. I also use 56v zener diodes on the screen taps to drop the screen voltage to around 300v.
I actually like the sound and flexibility of the tone control, a little boost here and a little cut there, makes life lot easier. The bass itself is quite deep and awesome without increasing cap size. Unless you have leakage from the coupling caps, there's no need to change to fancy modern caps. They sound just fine. (I like coloration and I am sure the "neutrality" fascists will be all over me but I couldn't care less. Your mileage may vary.)
I have several of these amps so when I have the time I would like to use a pair to do a mod with completely different circuit. One day I would like to implement the Pilot 232 circuit on this amp.
Oh, I also change the 6AU6 to triode connection with a 3.3k cathode resistor and it sounds great, albeit with less gain.
I tried the phono input using two halves of the 6EU7 but the hum is too loud to be usable due to the AC filament. The phono sounds is quite nice but the hum is too loud. You need DC filament to be quiet enough. I just don't use it right now for this mod.
I also added 1kΩ grid resistors for all the tubes to avoid oscillation.
--------------------------------------------------------------------
Below are stock schematics of the two amps for comparison.
AA-161
An externally hosted image should be here but it was not working when we last tested it.
AA-13
Two things:
First I don't like the 6AN8 pentode input being returned to ground only through a pot viper. After some time and corrosion where does that leave the input-grid to say nothing of the rest of the tube?.
Second I'd like to say I did a 6AN8 concertina a while ago and I do like the outcome very much.
I didn't do the regular version as in your Heathkit amp but a boot-strapped version with higher gain, probably around 3 times higher or there abouts.
I don't see a lot of using this circuit even if the principle's been around since the fourties.
If I had a Heathkit amp I'd try it out and compare it with the original 6AN8 splitter to see what sounds the best. It shouldn't be too hard to do the rewiring and fit the few extra components.
Take a look at the circuit. It seems to run the 6AN8 pretty close to the Heathkit setup even if I supply it from a little higher voltage.
What do you think?
rgds,
/tri-comp
First I don't like the 6AN8 pentode input being returned to ground only through a pot viper. After some time and corrosion where does that leave the input-grid to say nothing of the rest of the tube?.
Second I'd like to say I did a 6AN8 concertina a while ago and I do like the outcome very much.
I didn't do the regular version as in your Heathkit amp but a boot-strapped version with higher gain, probably around 3 times higher or there abouts.
I don't see a lot of using this circuit even if the principle's been around since the fourties.
If I had a Heathkit amp I'd try it out and compare it with the original 6AN8 splitter to see what sounds the best. It shouldn't be too hard to do the rewiring and fit the few extra components.
Take a look at the circuit. It seems to run the 6AN8 pretty close to the Heathkit setup even if I supply it from a little higher voltage.
What do you think?
rgds,
/tri-comp
Last edited:
That forum thread still doesn't say what's wrong with the 6AN8 and compare to what? To me all the 9-pin pentode/triode combo tubes sound the same to me, whether 6AN8, 7199, 6U8, 6GH8, etc... Just listen to them and make up your own mind. All purists like to point the finger at the line stage or tone control. But the beauty of an integrated amp is the line stage working in tandem with the amp section in harmonious way -- just ask those EICO HF-81 owners. The sound can be better than the sum of its parts.
By the way, I owned the Dyn SCA-35 before. I prefer the Heath, simply because it has more real estate and allows easier mods and tweaks. All good amps, the outcome depends on how much time you want to spend on improving it or accepting what it is.
Enjoy your AA-13.
P.S. I do not like UL operation at high voltage, not for sonic reason but for tube lifespan reason. Designers should respect the ratings in the tube manual regarding screen-grid voltage. Eking out the last drop of wattage has nothing to do with sound quality. Go for quality not quantity. Personally I found ultra-linear can sound wonderful at low voltage and for a tube like 6BQ5, I prefer running it under 300V.
By the way, I owned the Dyn SCA-35 before. I prefer the Heath, simply because it has more real estate and allows easier mods and tweaks. All good amps, the outcome depends on how much time you want to spend on improving it or accepting what it is.
Enjoy your AA-13.
P.S. I do not like UL operation at high voltage, not for sonic reason but for tube lifespan reason. Designers should respect the ratings in the tube manual regarding screen-grid voltage. Eking out the last drop of wattage has nothing to do with sound quality. Go for quality not quantity. Personally I found ultra-linear can sound wonderful at low voltage and for a tube like 6BQ5, I prefer running it under 300V.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.